1. Field of the Invention
The present invention generally relates to apparatuses for generating tones by use of a waveform memory, and more particularly to an apparatus which generates tones such as tones of percussion instruments by using a waveform memory efficiently.
2. Prior Art
In a conventional apparatus for generating tones such as a rhythm performance apparatus, each waveform data corresponding to each percussion instrument (such as a drum and a cymbal) are pre-stored in a waveform memory and the stored waveform data are selectively read out from the waveform memory, whereby such apparatus can play the rhythm performance. Before such a process, a player actually plays the percussion instruments so as to obtain the tones of the percussion instruments (percussive tones), and the waveforms of the percussive tones are sampled and converted into amplitude data by use of an analog-to-digital converter. The amplitude data are used as the above waveform data. For example, the amplitude data of eight bits are obtained at each sampling point. Usually, such amplitude data of eight bits are stored in the waveform memory, the bit number of one word (the one word bit number) of which is eight bits. Hence, one amplitude data at one sampling points (hereinafter, referred to as one sampling data) are assigned to one word storing area of the waveform memory.
In the above apparatus, the one word bit number of the waveform memory must be increased when the bit number of one sampling point (one sampling bit number) is increased. In order to improve the fidelity corresponding to the percussive tone source, it is necessary to increase the one sampling bit number. In some cases, it is preferable that one sampling bit number is set to twelve bits, for example. In this case, it is required to use a waveform memory in which the one word bit number is set to more than twelve bits. Hence, the conventional tone generating apparatus suffers a problem in that the cost thereof must be raised.
In order to solve the above problem, it can be considered to employ a method in which one sampling data are assigned to two-word storing area of the waveform memory. In this case, one sampling bit number is eight bits and the two-word bit number of the waveform memory is equal to sixteen bits (because one word bit number is eight bits), hence, four-bit storing area is remained not to be used at each two-word storing area of the waveform memory. Therefore, this method is disadvantageous in that an utilization efficiency of the waveform memory is relatively low.
In addition, specific waveform memory portion must be provided for storing noisy tones in the conventional waveform memory because plural waveform data are stored for each musical instrument. Hence, the conventional apparatus suffers another problem in that the storage capacity of the waveform memory must be increased.